Stumbling and Mumbling

Credibility and public borrowing

I had hoped to avoid the Budget, being unable to improve upon the mighty Mash's coverage. But I had the misfortune to hear Darling on the Today programme, which reminded me of why I so loathe the whole pantomime. It's a perversion of economics.
Darling was repeatedly asked: why should we believe your forecasts for public borrowing? His answer was pretty much wibble.
But the thing is, there's an answer which is both consistent with reasonable economics, and which defends him. It runs roughly thus:

You shouldn't believe my forecasts. Forecasts for public borrowing have always been subject to huge margins of error. There's a rule of thumb, which dates back at least as far as Nigella's dad, which says that the average error is 1% of GDP for each year of forecasting horizon - for both Treasury or private sector forecasts. That means my forecast for net borrowing of £43bn next year (pdf) should read £28-58bn, and the £38bn for 2009-10 should read £8-68bn, with a roughly one-third chance of borrowing being outside these ranges.
One reason for such big margins of error is that the government  isn't much in control of the public finances. Government borrowing is, by definition, the counterpart of private sector lending (Table I of this pdf). And this has recently been much higher than could have been expected, in two senses. First, company profits have persistently been higher than capital spending. Second, fast-growing Asian economies have generated high savings, which - because of asset shortages in the region - have been invested in developed economies such as ours; balance of payments figures (table J of this pdf) show that  foreign buying of UK debt (not just gilts) have been a record levels in recent years.
If someone's lending, someone's got to borrow. That's the government.
This flood of savings explains an important  fact - that despite the prospect of continued public borrowing, gilt yields are very low; index-linked yields are below 1%. Debt markets, then, are acting as if there's no problem with the public finances. And if the people who are stumping up the money don't seem to be worried about public borrowing, why should you be?

This answer, of course, raises other questions - not least about the fact that borrowing is only ever deferred taxation. But it answers the question about the credibility of borrowing forecasts.
But the thing is, Darling didn't say anything like this. He preferred to give the impression that he's on top of things, rather than give an economically coherent answer.
Which is why I hate Budgets. They are part of a folie a deux, in which both Chancellors and their media interlocuters pretend that it's possible to manipulate the public finances in detail. But the truth is that Chancellors don't have such control. And, with gilt yields so low, it doesn't - for now at least - much matter.

Sex and substitutes

Are goods complements or substitutes? This is a standard question in economics. And thanks to this survey, we can answer it for sex; is impersonating a politician a substitute or complement for sex with another person?
In theory it could be either. If we're not getting enough, we might take matters into our own hand - in which case the two are substitutes. But if sex is addictive, the two might be complements; more sex leads to more self-pleasuring.
Alas, the data is ambiguous. If we exclude Nigeria, where they get lots of sex and masturbate a little, the correlation between getting as much sex as you'd like and masturbating weekly is an insignificant minus 0.07. For example, only 14% of Japanese get as much sex as they'd like, but the proportion of them who self-pleasure weekly is about the same as do so in Mexico, where 56% get as much as they like.
The British, as you might imagine, score well for sex with Mrs Thumb and her four daughters but not so well for sex with others.
Now, you can quibble with the data. It doesn't distinguish here between men and women. And there might be differences in social norms which make people in some nations more reluctant than in others to own up to doing it with themselves but not reluctant to admit to disappointing sex lives.
These (perhaps important) doubts aside, this correlation implies that the cross-price elasticity of demand for masturbation with respect to sex should be roughly zero. Things that raise the price of sex won't much affect demand for masturbation, or vice versa. So make what you want of this.

Moral hazard and social discount rates

Why was Mervyn King - along with other economists - more worried (pdf) than Alastair Darling (a successor to Nigella’s dad)  about the moral hazard problem associated with bailing out Northern Rock and giving support to longer-term money markets?
Put it this way. Bailing out a bank today encourages banks to lend recklessly in future, in the belief they’ll be bailed out too. This means there’s a greater danger in future of a financial collapse.
Why does (did?) this prospect loom larger to King than Darling? It’s not just that King is more puritanical than Darling and instinctively hostile to borrowing - not at all. Nor is it, probably, that King is more aware of the effects of bad incentives and so attaches a higher probability to future financial troubles. Instead,  I suspect one overlooked reason lies in their time discount rates.
As a product of Cambridge in the 1960s, King will have been heavily influenced by the views of Pigou and Ramsey, that policy-makers should have a zero time-discount rate, because future people have the same moral status as current ones.
This means that he attaches (attached?) a high weight to the future cost of a probable financial collapse.
Darling, by contrast, attaches less weight to it. He’ll be out of office then, but faces the political costs of depositors’ distress today. So, even if both men were equally aware of the moral hazard problem, Darling would discount it more heavily, thus causing a difference between the two.
Which raises questions. Even if we take the moral hazard problem seriously - as we should - is King right to not discount rate it much? There are arguments (pdf) for policy-makers to discount the future.
And, is the government’s time discount rate consistent across policies? When I look at the bail-out of Northern Rock, I see a high time discount rate. When I look at policy towards climate change, I see a lower time discount rate. Am I right to see this? Am I right to think it inconsistent?

Inherited optimism bias

Why do Americans invest in English football clubs when Englishmen know it's money down the drain? Why is the US prone to bubbles of the sort described in Daniel Gross's Pop!? What possesses Martha Stewart to think she can compete with Nigella?
Here's a theory. These questions have a common answer. Americans, or at least white Americans, are more optimistic than other folk. This is because they have been selected to be so. To get in a boat and sail 3000 miles in search of a new life requires optimism - a belief that life will get better if you try. Fatalists, who reckon life's just a vale of tears and there's nowt you can do about it, wouldn't bother. And as parents pass such optimism down to their kids - through upbringing more than genes - so a nation of immigrants will be more optimistic than we old European nations.
"C'est la vie" and "mustn't grumble" are European sayings, not American ones.
The upshot is that Americans are more entrepreneurial, more willing to invest in new things, and willing to take on prospects with low odds of success than others.
Of course, there are exceptions. The greatest American of them all sang "Nothing's ever gonna be alright nohow". But he was influenced by blacks, who learnt from centuries to be pessimistic.
So, here's a question. Might this partially explain the US's racial divisions? Could it be that whites are better educated, more active in politics, more likely to set up businesses, and richer than blacks in part because they have inherited an optimism bias?
Or does this just show that any idiot can do evolutionary psychology?   

Why you should buy my book

Observant readers might have spotted a subtle plug for my book. Here are a few reasons why you should buy it:
1. It's not too long. It obeys Mahalanobis' rule, that books should be less than 300 pages.
2. You can read it as a toilet book, dipping in from time to time - though it also has a coherent linear narrative if you like a beginning, middle and end in that order.
3. There's almost no explicit class hatred; I've hidden the "chippiness."
4. There aren't that many references to Foucault or Feyerabend.
5. A former Chancellor is referred to only as Nigella's dad.
6. Surely, you want my book to outsell this, don't you?
7. The only reference to John Prescott's knob is in the first paragraph.
8. If it sells well, it'll show that blogging can be a platform for writers to enter the dead tree industry.
9. Who can dislike a book which quotes John Stuart Mill and the Blue Sky Boys on the same page?
10. There are no equations, except in the footnotes.
11. It's not mainly an anti-managerialist rant. It's packed with facts, some of them correct.
12. It's not too bad, if you like that sort of thing.

Wisdom in crowds?

Just how wise are crowds? We might be about to find out the hard way.
Todays figures (table I of this pdf) show that net household borrowing - borrowing minus savings - hit a record high of £10.15bn in Q4.
Granted, this is partly payback for low borrowing earlier in the year. Even so, net borrowing in 2006 was 2.1% of GDP. Though less than in 2004, this is comparable to what we saw at the peak of Nigella's dad's boom in 1988.
There are two flatly contradictory ways of interpreting this.
1. It's a vote of confidence in the future. Consumers are borrowing because they anticipate higher incomes in future. This is one implication of the permanent income hypothesis and wisdom of crowds thesis.
2. It's a sign that consumers are feckless and irresponsible. When the bills come in, they'll have to curb spending, possibly causing recession and falling house prices.
The empirical evidence here is ambiguous. High borrowing in 1988-89 did indeed lead to falling spending and house prices, and the improvement in household finances in the early 90s led to stronger spending.
However, before 1988 and since the mid-90s, the link between net borrowing and subsequent moves in spending or house prices has been weak.
We're therefore thrown  back onto our  prejudices about the wisdom of crowds.
And a big issue here is: are individuals' irrationalities correlated or not?
The wisdom of crowds hypothesis assumes they are not, so for each spendthrift, there's a miser. Which raises the question: what sort of mechanisms can cause irrationally high spending to become correlated across people? How strong are they?
In this sense, macroeconomic forecasting is an issue in crowd psychology.